Why operational diligence is non-optional
A well-prepared seller hands buyers a data room of audited financials, product mix, capex history. Useful, and insufficient. The data room tells you what the plant said it was. The floor tells you what it is. We have seen ~70 container glass diligences. Patterns repeat.
1. Campaign-life and refractory condition
Furnace campaign-life remaining vs the seller's 'normalised' assumption. Sidewall AZS thickness, crown sag, throat condition. A 12-month earlier rebuild than assumed is a $30M+ valuation-affecting finding.
2. KPI baseline and verification
Pack-to-melt, OEE, percent-pack, MTBF — all need to be replayed against raw production data, not the management deck. We have seen 8-point overstatements where 'planned' downtime was creatively excluded.
3. Management capability assessment
Leader Standard Work in place? Tier meetings running? KPI ownership clear? Most plants fail this — and the gap shows up as flat performance for 18 months post-close.
4. Mould inventory and SKU library state
Mould inventory is often a hidden asset/liability. SKU library state predicts the difficulty of any operational standardisation post-close.
5. Hot-end coating spec drift
Hot-end coating dose drift is invisible until cold-end rejects climb. We measure it directly during diligence.
6. Environmental compliance tail-risk
Local permit limits, NOx and particulate compliance, water effluent discharge limits. Sometimes a buyer inherits a non-compliance trajectory the seller didn't disclose.
7. 100-day post-close plan
Operational findings should walk into post-close execution without rework. Recommendations that don't survive into a 100-day plan were not recommendations — they were decoration.